Date Published: April 12, 2019
Publisher: Public Library of Science
Author(s): Xia Chaoxu, Nie Gaozhong, Fan Xiwei, Zhou Junxue, Pang Xiaoke, Francisco Martínez-Álvarez.
After an earthquake, the important task of emergency rescue work is to minimize casualties, but due to the suddenness of earthquake disasters, it is difficult to obtain enough disaster information immediately, especially personnel distribution and movement information. The traditional methods of obtaining disaster data are through reports from the disaster area or field investigations by the emergency rescue team; this work lags, and its efficiency is low. This paper analyzes the feasibility of using mobile phone location signal data in earthquake emergency rescue work in several respects, such as quantity, location, change rate, and epicentral distance. The results show that mobile phone location signal data can quickly obtain the situation of personnel distribution and quantity after an earthquake, and we find the change rate, distance, etc., can determine the approximate range of the earthquake impact field. Through the data distribution in different time periods, the movement of personnel after the earthquake can be obtained. Based on several situations, we can determine the basic situation of the disaster-stricken areas in times after the earthquake, especially the personnel relevant to the situation, and these data can provide a scientific basis for emergency rescue decision making.
In recent years, mobile phones have become a pervasive technology with users carrying them at almost all times. From this perspective, the ubiquity of such platforms has transformed mobile phones into one of the main sensors of human behavior. With the advent of the era of big data, mobile phones have become increasingly popular; the result is that people’s dependence on mobile phones has become increasingly high. Many users carry mobile phones at any time all day long. The GPS information from the mobile phone itself and the location information obtained in the process of using the mobile phone can reflect the real-time location information of the user. From this point of view, mobile phone location signal data can be seen as indicators of human activity. This ‘geographical footprint’ is similar to that collected when a cell phone call or text message is sent. Thus, social activity with GPS coordinates can be aggregated to and summarized by various spatial/political units for different analytical purposes[1–4]. Mobile phone-related data have also been widely used in different disciplines, such as research on the movement of people, the distribution of people after natural disasters, and the spread of epidemics.
The number of mobile phone location signals is dynamically changing. The movement of people mainly affects the location of mobile phone signal data. In general, signal data represent a person. After an earthquake occurs, the number of mobile phone signals in a certain area may change, so the change rate in the number of signals can reflect the movement of people in different regions. The formula for calculating the change rate is as follows:
where Na is the signal number of the previous stage and Nb is the signal number of the later stage. When α<0, the number of signals in the later phase is less than that in the previous phase. Conversely, when α> is 0, the number of signals in the later phase is greater than that in the previous phase.
We performed a detailed analysis of the results from three aspects. As shown in Fig 1, the change trend in the mobile phone signal data during the same time period was relatively stable on the day before the earthquake occurred (August 7, 2017). There was no significant fluctuation, especially at the time corresponding to the earthquake (21:19). The number of signals per minute was concentrated in the range of 200–500. From 19:00, the number of signals increased. The trend reached the maximum range at approximately 22:00; then, the number of signals began to decrease, showing a downward trend, but the rate of decrease was relatively stable, with no obvious transient changes.
In this paper, we present preliminary results to characterize the social response to an earthquake. We used the Jiuzhaigou earthquake as an example; we obtained mobile phone location signal data 2 hours before the earthquake and 8 hours after the earthquake in Jiuzhaigou and determined the behavior patterns of the people after the earthquake based on the distribution of and changes in the mobile phone location signal data. In contrast, we also collected data for the same period on the day before the earthquake. The distribution of personnel and the regular pattern of activities provide data support and a theoretical basis for earthquake emergency decision making. Our results show that the total amount of signal data in the study area can be divided into several stages:
This paper analyzes the feasibility of using mobile phone location signal data in earthquake emergency rescue work from several aspects, such as quantity, location, change rate, and epicentral distance. The results show that the mobile phone location signal data can quickly obtain the situation of the personnel distribution and quantity after the earthquake, and we find the change rate, the distance, etc. can determine the approximate range of the earthquake impact field. The change in data volume can well reflect the postearthquake movements of personnel. Through the change rate of data volume, people’s flow trend can be obtained. We can obtain the concentrated distribution area of the post-earthquake personnel by the rate of change of the mobile phone signal data in the overlapping area. And it is also possible to judge the flow direction of the personnel after the earthquake due to the rate of change of the signal data of the mobile phone in the non-overlapping region. After the earthquake, people mainly concentrated in urban areas, and the distribution location is relatively concentrated, mainly concentrated in the Jiuzhaigou County and Songpan County. The flow direction of the personnel after the earthquake is mainly from the Jiuzhaigou county towns located in the area of intensity of IX and VIII to the Songpan County and other areas of the intensity of VI and VII.